US8818098B2 - Apparatus and method for recognizing characters using a camera - Google Patents

Apparatus and method for recognizing characters using a camera Download PDF

Info

Publication number
US8818098B2
US8818098B2 US14069996 US201314069996A US8818098B2 US 8818098 B2 US8818098 B2 US 8818098B2 US 14069996 US14069996 US 14069996 US 201314069996 A US201314069996 A US 201314069996A US 8818098 B2 US8818098 B2 US 8818098B2
Authority
US
Grant status
Grant
Patent type
Prior art keywords
text
characters
character
store
store name
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
US14069996
Other versions
US20140055643A1 (en )
Inventor
Seong-taek Hwang
Sang-Wook Oh
Hyun-Soo Kim
Sang-Ho Kim
Hee-min Kwon
Yun-Je Oh
Byung-Jik Kim
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Samsung Electronics Co Ltd
Original Assignee
Samsung Electronics Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/222Studio circuitry; Studio devices; Studio equipment ; Cameras comprising an electronic image sensor, e.g. digital cameras, video cameras, TV cameras, video cameras, camcorders, webcams, camera modules for embedding in other devices, e.g. mobile phones, computers or vehicles
    • H04N5/225Television cameras ; Cameras comprising an electronic image sensor, e.g. digital cameras, video cameras, video cameras, camcorders, webcams, camera modules for embedding in other devices, e.g. mobile phones, computers or vehicles
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/20Image acquisition
    • G06K9/22Image acquisition using hand-held instruments
    • G06K9/228Hand-held scanners; Optical wands
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/62Methods or arrangements for recognition using electronic means
    • G06K9/72Methods or arrangements for recognition using electronic means using context analysis based on the provisionally recognised identity of a number of successive patterns, e.g. a word
    • G06K9/723Lexical context
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K2209/00Indexing scheme relating to methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K2209/01Character recognition

Abstract

A method is provided for recognizing characters of an image captured using a camera in a mobile terminal. The camera captures an image of a signboard. A text area is extracted from the captured image. Characters are recognized from the extracted text area. At least one text is generated by combining the recognized characters. The at least one text is compared with at least one store name within a predetermined range. A comparison result is output.

Description

PRIORITY

This application is a Continuation application of U.S. patent application Ser. No. 12/632,118, filed in the USPTO on Dec. 7, 2009, which claims priority under 35 U.S.C. §119(a) to a Korean Patent Application filed in the Korean Intellectual Property Office on Dec. 8, 2008 and assigned Serial No. 10-2008-0123957, the entire disclosure of which is incorporated herein by reference.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates generally to increasing a character recognition rate of an image received through a camera, and more particularly, to an apparatus and method for increasing a character recognition rate by extracting store names within a predetermined radius using a location information system and comparing the extracted store names with character information received through a camera of a mobile phone.

2. Description of the Related Art

With the increased popularity of mobile phones equipped with a camera, a variety of different service scenarios are being developed. Particularly, work is currently being made for a service in which a store name is recognized by the camera of the mobile phone and additional information related to the store name is then provided to a user of the mobile phone. Additionally, another service for use while traveling recognizes a signboard and translates the signboard for the traveler.

FIG. 1 is a flowchart illustrating a conventional operation for recognizing characters from a signboard in a mobile phone.

Referring to FIG. 1, a user captures an intended signboard using a camera of the mobile phone in step 101. In step 103, a text area is extracted from the captured image and converted to a black and white binary image. The binary text is segmented on a character basis in step 105, and distortion, such as noise, is compensated for in each character so that the character can be recognized normally in step 107. In step 109, each character is recognized through a character recognizer usually by best matching, exact matching, etc.

In step 111, to verify whether the text obtained by combining the compensated characters has been recognized successfully, it is determined whether the text is included in a database. Commonly, the database includes a dictionary function for determining whether the text has been recognized correctly. If the text is included in the database, the recognition result is displayed on an output portion of the mobile phone in step 113. Therefore, the user may search for related additional information. However, when the text is not included in the database, the user is notified that no valid text has been recognized in step 115.

When an intended signboard is captured by the camera of the mobile phone, it is not easy to analyze the captured image and recognize a text included in the captured image because various fonts and background images or colors that are available to signboards. Further, even from different images of the same signboard, the recognition rate of the same text from the signboard differs depending on lighting and a capturing angle. Further, if the text included in a signboard is a store name, the store name is often a proper noun in most cases, and therefore, it will not be recognized using the above-described dictionary function. Consequently, store names are difficult to recognize.

FIGS. 2A and 2B illustrate a conventional character recognition order. For example, referring to FIG. 2A, when a store name “

Figure US08818098-20140826-P00001
” is recognized, the store name is divided on a character basis and best matching is applied to the individual characters, thus producing a recognition result. For each character, character candidates with first to fifth priority levels are extracted and only characters with the highest priority levels are selected as an output recognition results. While this technique may lead to an accurate recognition result, a wrong recognition result may be obtained according to the angle and lighting of image capturing, like “
Figure US08818098-20140826-P00002
” illustrated in FIG. 2A.

Referring to FIG. 2B, when a store name “ebook” is captured, the store name is divided on a character basis and best matching is applied to the individual characters, thus producing a recognition result. For each character, character candidates with first to fifth priority levels are extracted and only characters with the highest priority levels are selected as output recognition results. While this technique may lead to an accurate recognition result, a wrong recognition result may be obtained according to the angle and lighting of image capturing, like “fboek” illustrated in FIG. 2B. Currently, there is no way to search for accurate additional information with the wrong recognition result.

SUMMARY OF THE INVENTION

The present invention has been designed to address at least the problems and/or disadvantages above, and to provide at least the advantages described below.

Accordingly, an aspect of the present invention is to provide an apparatus and method for quickly and accurately recognizing characters by extracting only store names within a predetermined radius around a user from a nationwide store name database using location information received, e.g., from a Global Positioning System (GPS), and comparing the extracted store names with character information included in a captured image of a signboard.

In accordance with an aspect of the present invention, a method is provided for recognizing characters of an image captured using a camera in a mobile terminal. The camera captures an image of a signboard. A text area is extracted from the captured image. Characters are recognized from the extracted text area. At least one text is generated by combining the recognized characters. The at least one text is compared with at least one store name within a predetermined range. A comparison result is output.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other aspects, features, and advantages of certain embodiments of the present invention will be more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a flowchart illustrating a conventional operation for recognizing characters from a signboard in a mobile phone;

FIGS. 2A and 2B illustrate a conventional character recognition order;

FIG. 3 is a block diagram of a mobile phone for recognizing a signboard according to an embodiment of the present invention;

FIG. 4 is a flowchart illustrating an operation for recognizing characters included in a signboard by a GPS according to an embodiment of the present invention;

FIG. 5 is a flowchart illustrating an operation for selecting text similar to text included in a signboard according to an embodiment of the present invention;

FIG. 6 is a flowchart illustrating an operation for comparing store names included in a database with text included in a signboard on a character basis according to an embodiment of the present invention; and

FIG. 7 is a flowchart illustrating an operation for comparing store names included in a database with text included in a signboard on a character basis according to another embodiment of the present invention.

Throughout the drawings, the same drawing reference numerals will be understood to refer to the same elements, features, and structures.

DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

The matters defined in the description such as a detailed construction and elements are provided to assist in a comprehensive understanding of certain embodiments of the present invention. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present invention. Also, descriptions of well-known functions and constructions are omitted for clarity and conciseness.

Conventionally, to recognize a text, a combination range is set for each character included in the text and extracted characters are combined. Therefore, if the text recognition is not accurate, a wrong recognition result is produced. To solve this problem, in accordance with an embodiment of the present invention, an apparatus and method are provided for accurately recognizing text by comparing store names, which are included in location information that is based on a current position of the mobile terminal received from, e.g., a Global Positioning System (GPS), with text of a captured image.

FIG. 3 is a block diagram of a mobile phone for recognizing a signboard according to an embodiment of the present invention. Referring to FIG. 3, the mobile phone includes a capturer 301, a pre-processor 303, a recognizer 305, and a post-processor 307.

The capturer 301 captures a store name written on an intended signboard using a camera of the mobile phone. The pre-processor 303 extracts a text area from the image captured by the capturer 301, converts the text area to a binary black and white image, and compensates for distortion of the binary back and white image. The recognizer 305 recognizes the compensated binary text area on a character-by-character basis. The post-processor 307 determines whether text obtained by combining the characters recognized by the recognizer 305 is accurate by referring to a DataBase (DB), and corrects the text, if necessary. The post-processor 307 includes a combination range setter 31, a character combiner 32, a display 33, a DB comparator 34, a DB classifier 35, and a location information acquirer 36.

The combination range setter 31 extracts recognition results of up to a predetermined priority level for each character according to a weight of the character. The priority level may be predetermined by a manufacturer or a user. The character combiner 32 combines characters according to the recognition results extracted by the combination range setter 31. The character combiner 32 produces an accurate character combination in conjunction with the DB comparator 34.

The location information acquirer 36 acquires location information of an area within a predetermined range from a current position of the mobile phone and acquires additional information corresponding to the location information. For example, the current position of the mobile terminal can be provided using GPS. The location information includes all store names within the predetermined range. The DB classifier 35 separates the store names from the location information received from the location information acquirer 36 and stores the store names. Additionally, the additional information corresponding to the store names may also be separated and stored. The DB comparator 34 compares text obtained by combining the characters in the character combiner 32 with at least one name acquired from the DB classifier 35.

FIG. 4 is a flowchart illustrating an operation for recognizing text included in a signboard according to an embodiment of the present invention.

Referring to FIG. 4, a user captures an image of an intended signboard using a camera included in a mobile phone in step 401. In step 403, a text area is extracted from the captured image and converted to a binary black and white image to facilitate text recognition. The binary text is divided into characters in step 405 and the distortion of each character, such as noise, is compensated for in step 407, in order to accurately recognize the character. In step 409, each character is recognized by a character recognizer.

In step 411, location information of an area within a predetermined range from a current location of the mobile phone is acquired. For example, the current position of the mobile terminal can be provided using GPS. The location information includes store names within the predetermined range. For example, if a current location of the mobile phone on the campus of San Jose State University, the location information will include store names around San Jose State University.

In step 413, the store names are classified and stored in the mobile terminal. For example, all store names included in the location information may be extracted and stored, or only store names having as many characters as in the captured signboard may be extracted and stored.

In step 415, the mobile phone sets character combination ranges, i.e., recognition results of up to a predetermined priority level are extracted for each recognized character according to a weight of the character. For example, if six characters are recognized and a character combination range for each character is up to a fifth priority level, i.e., characters of first to fifth priority levels for each recognized character, 30 weighted characters may be extracted (for illustration, characters of up to a fifth priority level are extracted for each character in FIGS. 2A and 2B).

In step 417, characters are combined to text in a descending order of weights according to the combination range. If the combination range of each character is first to fifth priority levels, the number of the resulting combined texts is 15,625.

Each text is compared with the stored store names in step 419. If the text does not match any of the store names stored in the DB and there are still combinations to try in step 421, steps 417 and 419 are repeated within a range that does not exceed the combination range. If there is no text matching any of the store names stored in the DB despite completed character combining in step 421, the text obtained by combining characters with the highest priority levels is displayed on an output portion of the mobile phone, notifying that the text is not in the DB in step 425. However, upon detection of a text matching a store name stored in the DB in step 419, the text is displayed on the output portion of the mobile phone in step 423.

FIG. 5 is a flowchart illustrating an operation for selecting a text similar to a text included in a signboard according to an embodiment of the present invention. More specifically, while two recognition results are output in FIG. 4, in FIG. 5 more character combinations are output to the user at one time, e.g., in a list, such that a user may select a combination.

Referring to FIG. 5, the user captures an image of an intended signboard using a camera included in a mobile phone in step 501. In step 503, a text area is extracted from the captured image and converted to a binary black and white image to facilitate text recognition. The binary text is divided into characters in step 505 and the distortion of each character such as noise is compensated for in step 507, in order to accurately recognize the character. In step 509, each character is recognized by a character recognizer.

In step 511, location information of an area within a predetermined range from a current location of the mobile phone is acquired. For example, the current position of the mobile terminal can be provided using GPS. The location information includes store names within the predetermined range.

In step 513, the store names are classified and stored in the mobile terminal. For example, all store names included in the location information may be extracted and stored, or only store names having as many characters as in the captured signboard may be extracted and stored.

In step 515, the mobile phone sets character combination ranges, i.e., recognition results of up to a predetermined priority level are extracted for each recognized character according to a weight of the character. In step 517, characters are combined to text in a descending order of weights according to the combination ranges.

Store names similar to the generated text are classified and output sequentially by comparing the store names stored in the DB with the generated text in step 519. For example, to classify a store name as similar to the generated text, the store name is compared with the text on a character-by-character basis and if the store name matches to text by a predetermined number of characters, the store name is determined to be a similar store name. The number of similar store names to be output is freely set by the manufacturer or the user. In step 521, the user may select an intended text from among a predetermined number of texts output to the output portion.

FIGS. 6 and 7 are flowcharts illustrating operations for comparing store names included in the DB with a text included in an image of a signboard according to embodiments of the present invention. More specifically, in FIG. 6, the comparison begins with the first character of a recognized text, whereas in FIG. 7, the comparison begins with the last character of the recognized text.

In FIG. 6, steps 601 to 617 are equivalent to steps 501 and 517 of FIG. 5, which were described above. Accordingly, a description of these steps will not be repeated below.

Referring to FIG. 6, in step 619, the mobile phone determines whether the first character of a specific text matches the first character of any store name included in the DB. If the first character of the text does not match the first character of any store name included in the DB and there are still possible first character combinations in step 621, steps 617 and 619 are repeated until a text having the first character matching the first character of any store name in the DB is detected, or until there are no more first character combinations available in step 621. If none of the first characters of the texts match any of the first characters of the store names included in the DB and there are no more first character combinations available in step 621, a message indicating no store name matching the text is output on the display in step 629.

When there is at least one store name having the first characters matching the first characters of texts in step 619, the store names and the texts are re-classified in step 623. For example, if the first characters of the texts are “

Figure US08818098-20140826-P00003
”, “
Figure US08818098-20140826-P00004
” and “
Figure US08818098-20140826-P00005
”, these characters are compared with the first characters of store names stored in the DB and store names each having the first character being “
Figure US08818098-20140826-P00006
”, “
Figure US08818098-20140826-P00007
” or “
Figure US08818098-20140826-P00008
” are re-classified. If there is no store name having the first character being “
Figure US08818098-20140826-P00009
”, only the other texts having their first characters being “
Figure US08818098-20140826-P00010
” or “
Figure US08818098-20140826-P00011
” are re-classified. The next character within the re-classified combination range is compared with the next character of the re-classified store names in step 625 and the comparison results are re-classified.

As another example, if the first characters of the texts are “e”, “f” and “o”, these characters are compared with the first characters of store names stored in the DB and store names each having the first character being “e”, “f” or “o” are re-classified. If there is no store name having the first character being “o”, only the other texts having their first characters being “e” or “f” are re-classified. The next character within the re-classified combination range is compared with the next character of the re-classified store names and the comparison results are re-classified. When there is a store name matching the text in the first to last characters, the store name is displayed on the display in step 627.

In FIG. 7, steps 701 to 729 are equivalent to steps 601 and 629 of FIG. 6, which were described above, except that 719 starts with a last character instead of a first character as in step 619. Accordingly, a description of these steps will not be repeated below.

For example, referring to FIG. 7, when there is a store name having a last character matching a last character of a text, the store names and the texts are re-classified in step 723. That is, if the last characters of the texts are “

Figure US08818098-20140826-P00012
”, “
Figure US08818098-20140826-P00013
” and “
Figure US08818098-20140826-P00014
”, these characters are compared with the last characters of store names stored in the DB and store names each having the last character being “
Figure US08818098-20140826-P00015
”, “
Figure US08818098-20140826-P00016
” or “
Figure US08818098-20140826-P00017
”, are re-classified. If there is no store name having the last character being “
Figure US08818098-20140826-P00018
”, only the other texts having their last characters being “
Figure US08818098-20140826-P00019
” or “
Figure US08818098-20140826-P00020
” are re-classified. The second to last character within the re-classified combination range is compared with the second to last character of the re-classified store names in step 725 and the comparison results are re-classified.

As another example, the last characters of the texts are “k”, “h” and “l”, these characters are compared with the last characters of store names stored in the DB and store names each having the last character being “k”, “h” or “l”, are re-classified. If there is no store name having the last character being “l”, only the other texts having their last characters being “k” or “h” are re-classified. The second to last character within the re-classified combination range is then compared with the second to last character of the re-classified store names and the comparison results are re-classified. When there is a store name matching a text in the last to first characters, the store name is displayed on the display in step 727.

The comparison and re-classification operations of FIGS. 6 and 7 reduce the number of comparisons. Further, if only one re-classified text remains before the first or last character comparison, it is also possible to display this directly on the display without performing another reclassification.

As is apparent from the above description above, the above-described embodiments of the present invention extract a text area from image of a signboard captured using a camera, and increase the recognition rate of the text area based on location information relating to a current position of the user. Therefore, information can be searched for fast and accurately.

While the present invention has been shown and described with reference to certain embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the appended claims and their equivalents.

Claims (13)

What is claimed is:
1. A method for recognizing characters of an image captured using a camera in a mobile terminal, comprising:
capturing, by the camera, an image of a signboard;
extracting a text area from the captured image;
recognizing characters from the extracted text area;
generating at least one text by combining the recognized characters;
comparing the at least one text with at least one store name within a predetermined range; and
outputting a comparison result.
2. The method of claim 1, wherein recognizing the characters from the extracted text area comprises:
converting the extracted text area to a binary black and white text area;
dividing the binary text area into the characters;
compensating for distortion of each of the characters; and
recognizing each of the compensated characters.
3. The method of claim 1, further comprising:
generating similar characters of up to a predetermined priority level for each of the recognized characters of the extracted text area;
acquiring location information of an area within a predetermined range from a current position of the mobile terminal; and
extracting at least one store name from the location information.
4. The method of claim 3, wherein generating the similar characters comprises generating similar results up to a predetermined priority level in a descending order of weights, for each of the characters.
5. The method of claim 3, further comprising:
classifying the at least one store name included in the location information, and storing the classified at least one store name in a database.
6. The method of claim 5, wherein storing the classified at least one store name comprises:
separating only store names having as many characters as the characters from the extracted text area; and
storing the separated store names.
7. The method of claim 1, wherein comparing the at least one text with the at least one store name comprises:
comparing each of the at least one text with the at least one store name, until a text matching any extracted store name is detected, and
wherein outputting the comparison result comprises:
outputting the text matching any extracted store name; and
outputting a text obtained by combining similar characters with highest weights on a character-by-character basis, if no text matches to any extracted store name.
8. The method of claim 1, wherein comparing the at least one text with the at least one store name comprises:
identifying store names similar to the at least one text, and
wherein outputting the comparison result comprises:
outputting similar store names, for user selection.
9. The method of claim 8, wherein identifying the store names similar to the at least one text comprises:
comparing characters of the at least one text with characters of the at least one store name; and
identifying store names that are identical to the at least one text in a predetermined number of characters.
10. The method of claim 3, wherein the current position of the mobile terminal is determined by a Global Positioning System (GPS).
11. The method of claim 1, wherein comparing the at least one text with the at least one store name comprises:
comparing each character included in each of the at least one text with each character included in each of the at least one store name in a predetermined order.
12. The method of claim 1, wherein comparing the at least one text with at least one store name comprises:
comparing a first character of each of the at least one text with a first character of each of the at least one extracted store name;
primarily re-classifying store names each having the first character identical to the first character of any of the at least one text;
comparing a second character of each of the re-classified store names with a second character of each of the at least one text;
secondarily re-classifying store names each having the second character identical to the second character of any of the at least one text among the re-classified store names; and
repeating the secondary re-classification, until comparison is completed over all characters included in the at least one text.
13. The method of claim 1, wherein comparing the at least one text with at least one store name comprises comparing each character included in each of the at least one text with each character included in each of the at least one extracted store name in a last-to-first character order.
US14069996 2008-12-08 2013-11-01 Apparatus and method for recognizing characters using a camera Active US8818098B2 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
KR20080123957A KR101035744B1 (en) 2008-12-08 2008-12-08 Apparatus and method for character recognition using camera
KR10-2008-0123957 2008-12-08
US12632118 US8594424B2 (en) 2008-12-08 2009-12-07 Apparatus and method for recognizing characters using a camera
US14069996 US8818098B2 (en) 2008-12-08 2013-11-01 Apparatus and method for recognizing characters using a camera

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US14069996 US8818098B2 (en) 2008-12-08 2013-11-01 Apparatus and method for recognizing characters using a camera

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US12632118 Continuation US8594424B2 (en) 2008-12-08 2009-12-07 Apparatus and method for recognizing characters using a camera

Publications (2)

Publication Number Publication Date
US20140055643A1 true US20140055643A1 (en) 2014-02-27
US8818098B2 true US8818098B2 (en) 2014-08-26

Family

ID=42230623

Family Applications (2)

Application Number Title Priority Date Filing Date
US12632118 Active 2032-09-23 US8594424B2 (en) 2008-12-08 2009-12-07 Apparatus and method for recognizing characters using a camera
US14069996 Active US8818098B2 (en) 2008-12-08 2013-11-01 Apparatus and method for recognizing characters using a camera

Family Applications Before (1)

Application Number Title Priority Date Filing Date
US12632118 Active 2032-09-23 US8594424B2 (en) 2008-12-08 2009-12-07 Apparatus and method for recognizing characters using a camera

Country Status (2)

Country Link
US (2) US8594424B2 (en)
KR (1) KR101035744B1 (en)

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101035744B1 (en) * 2008-12-08 2011-05-20 삼성전자주식회사 Apparatus and method for character recognition using camera
CN102150418B (en) * 2008-12-22 2014-10-15 松下电器产业株式会社 Image amplification apparatus, a method and an integrated circuit
WO2012161706A1 (en) * 2011-05-24 2012-11-29 Hewlett-Packard Development Company, L.P. Region of interest of an image
US9064191B2 (en) 2012-01-26 2015-06-23 Qualcomm Incorporated Lower modifier detection and extraction from devanagari text images to improve OCR performance
US9053361B2 (en) 2012-01-26 2015-06-09 Qualcomm Incorporated Identifying regions of text to merge in a natural image or video frame
US9076242B2 (en) 2012-07-19 2015-07-07 Qualcomm Incorporated Automatic correction of skew in natural images and video
US9014480B2 (en) 2012-07-19 2015-04-21 Qualcomm Incorporated Identifying a maximally stable extremal region (MSER) in an image by skipping comparison of pixels in the region
US9262699B2 (en) 2012-07-19 2016-02-16 Qualcomm Incorporated Method of handling complex variants of words through prefix-tree based decoding for Devanagiri OCR
US9047540B2 (en) 2012-07-19 2015-06-02 Qualcomm Incorporated Trellis based word decoder with reverse pass
US9141874B2 (en) 2012-07-19 2015-09-22 Qualcomm Incorporated Feature extraction and use with a probability density function (PDF) divergence metric
CN103020812B (en) * 2012-12-05 2016-08-24 上海合合信息科技发展有限公司 Convenient electronic professional electronic notebook and added to a calendar
US9405978B2 (en) 2013-06-10 2016-08-02 Globalfoundries Inc. Prioritization of facial recognition matches based on likely route
JP2015040908A (en) * 2013-08-20 2015-03-02 株式会社リコー Information processing apparatus, information update program, and information update method
US9747528B1 (en) * 2016-10-26 2017-08-29 International Business Machines Corporation Capturing contextual information on a device

Citations (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4850025A (en) * 1985-09-27 1989-07-18 Sony Corporation Character recognition system
US4903206A (en) * 1987-02-05 1990-02-20 International Business Machines Corporation Spelling error correcting system
KR950007888A (en) 1993-09-10 1995-04-15 죠셉 에프. 셔츠 Catheter assembly having a display surface
US5583978A (en) * 1994-05-27 1996-12-10 Bitstream Inc. Apparatuses and methods for creating and using portable fonted texts with embedded automatically-created font descriptions
US5825999A (en) * 1993-12-28 1998-10-20 Victor Company Of Japan, Ltd. Minimum character region determining apparatus and method, character string generating apparatus for generating a character string according to the minimum character region determining method
US5999647A (en) * 1995-04-21 1999-12-07 Matsushita Electric Industrial Co., Ltd. Character extraction apparatus for extracting character data from a text image
US6246794B1 (en) * 1995-12-13 2001-06-12 Hitachi, Ltd. Method of reading characters and method of reading postal addresses
US6266442B1 (en) * 1998-10-23 2001-07-24 Facet Technology Corp. Method and apparatus for identifying objects depicted in a videostream
US6341176B1 (en) * 1996-11-20 2002-01-22 Matsushita Electric Industrial Co., Ltd. Method and apparatus for character recognition
US20020154817A1 (en) * 2001-04-18 2002-10-24 Fujitsu Limited Apparatus for searching document images using a result of character recognition
US6473517B1 (en) * 1999-09-15 2002-10-29 Siemens Corporate Research, Inc. Character segmentation method for vehicle license plate recognition
US6640010B2 (en) * 1999-11-12 2003-10-28 Xerox Corporation Word-to-word selection on images
US20040057619A1 (en) * 2002-09-11 2004-03-25 Chae-Whan Lim Apparatus and method for recognizing a character image from an image screen
US6745161B1 (en) * 1999-09-17 2004-06-01 Discern Communications, Inc. System and method for incorporating concept-based retrieval within boolean search engines
US6834121B2 (en) * 1999-12-22 2004-12-21 Nec Corporation Apparatus for rough classification of words, method for rough classification of words, and record medium recording a control program thereof
US6870947B2 (en) * 2001-07-24 2005-03-22 Ncr Corporation Method of processing items in a check processing system and an apparatus therefor
US7010519B2 (en) * 2000-12-19 2006-03-07 Hitachi, Ltd. Method and system for expanding document retrieval information
US7031553B2 (en) * 2000-09-22 2006-04-18 Sri International Method and apparatus for recognizing text in an image sequence of scene imagery
US7069240B2 (en) * 2002-10-21 2006-06-27 Raphael Spero System and method for capture, storage and processing of receipts and related data
KR20060133430A (en) 2005-06-20 2006-12-26 삼성전자주식회사 Method and system for providing user with image related information and mobile communication system
KR20070099138A (en) 2006-04-03 2007-10-09 주식회사 오픈에스이 The examination system for the documents created with optical character recognition and the method thereof
US20080011841A1 (en) * 2005-02-03 2008-01-17 Yottamark, Inc. System and Method of Detecting Product Code Duplication and Product Diversion
KR100838580B1 (en) 2007-01-17 2008-06-20 한국과학기술원 System for providing store information using cell-phone, and the method therefor
US7606439B2 (en) * 2004-03-23 2009-10-20 Realeyes 3D Method for extracting raw data from an image resulting from a camera shot
US7616136B2 (en) * 2005-01-26 2009-11-10 Qisda Corporation Message compression methods and systems
US7970213B1 (en) * 2007-05-21 2011-06-28 A9.Com, Inc. Method and system for improving the recognition of text in an image
US8117458B2 (en) * 2006-05-24 2012-02-14 Vidoop Llc Methods and systems for graphical image authentication
US8290273B2 (en) * 2009-03-27 2012-10-16 Raytheon Bbn Technologies Corp. Multi-frame videotext recognition
US8335402B1 (en) * 2008-01-23 2012-12-18 A9.Com, Inc. Method and system for detecting and recognizing text in images
US8594424B2 (en) * 2008-12-08 2013-11-26 Samsung Electronics Co., Ltd Apparatus and method for recognizing characters using a camera

Patent Citations (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4850025A (en) * 1985-09-27 1989-07-18 Sony Corporation Character recognition system
US4903206A (en) * 1987-02-05 1990-02-20 International Business Machines Corporation Spelling error correcting system
KR950007888A (en) 1993-09-10 1995-04-15 죠셉 에프. 셔츠 Catheter assembly having a display surface
US5825999A (en) * 1993-12-28 1998-10-20 Victor Company Of Japan, Ltd. Minimum character region determining apparatus and method, character string generating apparatus for generating a character string according to the minimum character region determining method
US5583978A (en) * 1994-05-27 1996-12-10 Bitstream Inc. Apparatuses and methods for creating and using portable fonted texts with embedded automatically-created font descriptions
US5999647A (en) * 1995-04-21 1999-12-07 Matsushita Electric Industrial Co., Ltd. Character extraction apparatus for extracting character data from a text image
US6246794B1 (en) * 1995-12-13 2001-06-12 Hitachi, Ltd. Method of reading characters and method of reading postal addresses
US6341176B1 (en) * 1996-11-20 2002-01-22 Matsushita Electric Industrial Co., Ltd. Method and apparatus for character recognition
US6266442B1 (en) * 1998-10-23 2001-07-24 Facet Technology Corp. Method and apparatus for identifying objects depicted in a videostream
US6473517B1 (en) * 1999-09-15 2002-10-29 Siemens Corporate Research, Inc. Character segmentation method for vehicle license plate recognition
US6745161B1 (en) * 1999-09-17 2004-06-01 Discern Communications, Inc. System and method for incorporating concept-based retrieval within boolean search engines
US6640010B2 (en) * 1999-11-12 2003-10-28 Xerox Corporation Word-to-word selection on images
US6834121B2 (en) * 1999-12-22 2004-12-21 Nec Corporation Apparatus for rough classification of words, method for rough classification of words, and record medium recording a control program thereof
US7031553B2 (en) * 2000-09-22 2006-04-18 Sri International Method and apparatus for recognizing text in an image sequence of scene imagery
US7010519B2 (en) * 2000-12-19 2006-03-07 Hitachi, Ltd. Method and system for expanding document retrieval information
US20020154817A1 (en) * 2001-04-18 2002-10-24 Fujitsu Limited Apparatus for searching document images using a result of character recognition
US7142716B2 (en) * 2001-04-18 2006-11-28 Fujitsu Limited Apparatus for searching document images using a result of character recognition
US6870947B2 (en) * 2001-07-24 2005-03-22 Ncr Corporation Method of processing items in a check processing system and an apparatus therefor
US20040057619A1 (en) * 2002-09-11 2004-03-25 Chae-Whan Lim Apparatus and method for recognizing a character image from an image screen
US7069240B2 (en) * 2002-10-21 2006-06-27 Raphael Spero System and method for capture, storage and processing of receipts and related data
US7606439B2 (en) * 2004-03-23 2009-10-20 Realeyes 3D Method for extracting raw data from an image resulting from a camera shot
US7616136B2 (en) * 2005-01-26 2009-11-10 Qisda Corporation Message compression methods and systems
US20080011841A1 (en) * 2005-02-03 2008-01-17 Yottamark, Inc. System and Method of Detecting Product Code Duplication and Product Diversion
US20070036469A1 (en) 2005-06-20 2007-02-15 Samsung Electronics Co., Ltd. Method and system for providing image-related information to user, and mobile terminal therefor
KR20060133430A (en) 2005-06-20 2006-12-26 삼성전자주식회사 Method and system for providing user with image related information and mobile communication system
KR20070099138A (en) 2006-04-03 2007-10-09 주식회사 오픈에스이 The examination system for the documents created with optical character recognition and the method thereof
US8117458B2 (en) * 2006-05-24 2012-02-14 Vidoop Llc Methods and systems for graphical image authentication
KR100838580B1 (en) 2007-01-17 2008-06-20 한국과학기술원 System for providing store information using cell-phone, and the method therefor
US7970213B1 (en) * 2007-05-21 2011-06-28 A9.Com, Inc. Method and system for improving the recognition of text in an image
US8335402B1 (en) * 2008-01-23 2012-12-18 A9.Com, Inc. Method and system for detecting and recognizing text in images
US8594424B2 (en) * 2008-12-08 2013-11-26 Samsung Electronics Co., Ltd Apparatus and method for recognizing characters using a camera
US8290273B2 (en) * 2009-03-27 2012-10-16 Raytheon Bbn Technologies Corp. Multi-frame videotext recognition

Also Published As

Publication number Publication date Type
US8594424B2 (en) 2013-11-26 grant
KR101035744B1 (en) 2011-05-20 grant
US20140055643A1 (en) 2014-02-27 application
US20100141788A1 (en) 2010-06-10 application
KR20100065573A (en) 2010-06-17 application

Similar Documents

Publication Publication Date Title
Rodriguez et al. Local gradient histogram features for word spotting in unconstrained handwritten documents
US5917941A (en) Character segmentation technique with integrated word search for handwriting recognition
US20120092329A1 (en) Text-based 3d augmented reality
US20110066421A1 (en) User-interactive automatic translation device and method for mobile device
US20130195376A1 (en) Detecting and correcting skew in regions of text in natural images
US20090040215A1 (en) Interpreting Sign Language Gestures
US20050102139A1 (en) Information processing method and apparatus
US8194921B2 (en) Method, appartaus and computer program product for providing gesture analysis
US20090028435A1 (en) Character image extracting apparatus and character image extracting method
US20040006467A1 (en) Method of automatic language identification for multi-lingual text recognition
US20040057619A1 (en) Apparatus and method for recognizing a character image from an image screen
US20150235110A1 (en) Object recognition or detection based on verification tests
US7929771B2 (en) Apparatus and method for detecting a face
JP2002197103A (en) Method and system for providing data
US20080212837A1 (en) License plate recognition apparatus, license plate recognition method, and computer-readable storage medium
US20070136071A1 (en) Apparatus and method for speech segment detection and system for speech recognition
US20120114245A1 (en) Online Script Independent Recognition of Handwritten Sub-Word Units and Words
US7415165B2 (en) Red-eye detection device, red-eye detection method, and red-eye detection program
US20110222774A1 (en) Image feature detection based on application of multiple feature detectors
US20140168478A1 (en) Text Image Quality Based Feedback For Improving OCR
US20150356368A1 (en) Entrance detection from street-level imagery
US20130054240A1 (en) Apparatus and method for recognizing voice by using lip image
US20130058575A1 (en) Text detection using image regions
US20150347861A1 (en) Object-Of-Interest Detection And Recognition With Split, Full-Resolution Image Processing Pipeline
US20080226171A1 (en) Correcting device and method for perspective transformed document images

Legal Events

Date Code Title Description
MAFP

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551)

Year of fee payment: 4